## Colophon tags:: url:: https://michaeljhannon.substack.com/p/do-we-know-anything-about-politics %% title:: Do We Know Anything About Politics? type:: [[clipped-note]] author:: [[@michaeljhannon.substack.com]] %% ## Notes > Do We Know Anything About Politics? — [view in context](https://hyp.is/2IjdyAM1EfCrUDuxUcnsQw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) ⬆️ date:: [[2025-03-17]] tags:: [[%tie]] [[societies]] [[%t]]> In Political Humility: The Limits of Knowledge in Our Partisan Political Climate, Blake Roeber (a philosopher at the University of Notre Dame) argues that most of what we think we know about politics is not knowledge at all. “We have almost no political knowledge,” he claims. The reason is straightforward: our political beliefs are shaped almost entirely by testimony—information passed along by media, social networks, and partisan sources—that is often unreliable, biased, or misleading. What’s more, we rarely have the means to verify this information independently. This creates a pervasive illusion of knowledge, where confidence flourishes despite—and sometimes because of—ignorance. — [view in context](https://hyp.is/3gVc4AM1EfCIz98N2bZSVg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > While I agree with Roeber that our political confidence often exceeds our actual knowledge, I’m not convinced we should be as humble as he suggests. — [view in context](https://hyp.is/9VnjRgM3EfCfNAOhP7IQHQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Why can’t most of us know the answers to these questions? Roeber gives two main arguments. First, political issues are incredibly complex, requiring expertise in economics, public policy, history, and social science—far beyond what any individual can assess independently. Second, political partisanship itself clouds our judgment. — [view in context](https://hyp.is/F7Ch0gM4EfCpzOshq7HjxA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > However, we tend to lack knowledge of contested political issues, the kinds of questions that divide major parties and shape partisan debate. These include: Whether raising the minimum wage to $15 would help or hurt the economy.Whether stricter gun control laws reduce crime or make people less safe.Whether free trade agreements benefit the average worker or primarily enrich corporations.Whether higher taxes on the wealthy lead to economic growth through redistribution or stifle investment and job creation.Whether affirmative action policies improve long-term outcomes for marginalized groups or create unintended harms. — [view in context](https://hyp.is/F8StdgM4EfCspydfATbCsQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > As Roeber puts it, “We are simply incapable of thinking straight about political issues.” Once an issue becomes polarized, we no longer approach it as an open question but rather as a test of ideological loyalty. Our political thinking is plagued by too many psychological biases, making genuine knowledge nearly impossible to attain. — [view in context](https://hyp.is/bxdDNgM4EfCOsy-jdRPr8g/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > The central argument in Political Humility is admirably straightforward: We get our political information from testimony. (Defended in Chapter 2.)Political testimony is unreliable. (Defended in Chapter 3.)Therefore, we have little political knowledge. (Defended in Chapter 4.)On the basis of this argument, Roeber argues for a further conclusion, namely: “We should hold very few political beliefs.” — [view in context](https://hyp.is/jTs6AgM4EfCg0-f0Uo3nrg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) ⬆️ Worth noting that much of reportage today reflects this. Tbf, it is meant to be the "first draft". The challenge is that we don't go beyond that. > Political issues—such as whether raising the minimum wage reduces unemployment or whether climate policies effectively curb emissions—require specialized knowledge in economics, science, or geopolitics, none of which individuals can assess independently. As a result, we rely on news reports, expert analyses, conversations, and social media to form our beliefs. Even when we think we are reasoning for ourselves, we are usually synthesizing secondhand information. Crucially, Roeber emphasizes that this reliance on testimony is not just common but necessary. Given the vast scale of political knowledge, no one can verify all relevant facts firsthand — [view in context](https://hyp.is/VTHhhgM6EfCiUD-PstaMIQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > In defense of premise 2, Roeber argues that political testimony is systematically unreliable due to bias, misinformation, and social pressures. Politicians, media outlets, and advocacy groups have strong incentives to frame issues in ways that serve their interests rather than present objective facts, leading to selective reporting, exaggeration, and outright falsehoods — [view in context](https://hyp.is/3BLlGgM6EfCuayOMI2Kp9A/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > This leads to the conclusion that “it is impossible for any person to know the answers to more than just a handful of political questions.” If nearly all of our political beliefs are based on testimony (premise 1), and political testimony is systematically unreliable (premise 2), then our political beliefs are built on an epistemically unstable foundation. — [view in context](https://hyp.is/6W3BngM6EfCIFY-1f5HEgg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > I’ll suggest two ways to defend the idea that we can justifiably hold onto our political beliefs, even if Roeber is right that we lack political knowledge. — [view in context](https://hyp.is/esZdwgM7EfCgFc8MXneSeg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > If we interpret Roeber as only arguing that we lack knowledge, then his conclusion—that “we should hold very few political beliefs”—might be too strong. We can justifiably hold political beliefs even while recognizing that they do not meet the highest epistemic standard. — [view in context](https://hyp.is/qQgeUAM7EfCZf-NHQB3u8Q/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > What matters is whether those beliefs are responsibly formed, based on the evidence, and held with the appropriate degree of confidence. — [view in context](https://hyp.is/qRU69AM7EfCIphOv4TLzIg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > If so, then it’s perfectly appropriate to have political beliefs, as long as we remain aware of their epistemic limitations and avoid mistaking them for certain knowledge. This aligns with Roeber’s observation that people are often too confident in their political beliefs. — [view in context](https://hyp.is/hVsytgM8EfCwqbOQMoIuoA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Roeber could be making a stronger claim: not only do we lack political knowledge, but we also lack justification for most of our political beliefs. This position makes his argument more forceful. The reasoning would go as follows:We have little justification for our political beliefs.If a belief lacks justification, we should not hold it.Therefore, we should hold very few political beliefs.This represents a more radical form of skepticism — [view in context](https://hyp.is/k3yomAM8EfCpNm-4c85o4A/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > To illustrate, consider reliabilist theories of justification, which hold that a belief is justified only if formed through a process that reliably tracks the truth. (This view is popular in epistemology.) Roeber argues that our political beliefs fail this test because they are shaped by unreliable mechanisms—biased media, partisan narratives, social pressures, and cognitive biases like confirmation bias and motivated reasoning. — [view in context](https://hyp.is/E5UiogM-EfCt7PvUoDSosw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Moreover, since most people lack the expertise to independently verify political claims, they have no reliable way to correct these distortions. — [view in context](https://hyp.is/GxSwkgM-EfCYQAfK_ou4IA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Unlike scientific beliefs, which aim primarily at truth-tracking, political beliefs often serve non-epistemic goals—fostering social cohesion, signaling group identity, and reinforcing moral commitments. For many, political beliefs are deeply tied to personal values and community belonging, shaping how they relate to others and engage with the world. — [view in context](https://hyp.is/NyWvjgM-EfCrcuvMymUiWw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > this brings me to the second reason why Roeber’s argument may not require people to abandon their political beliefs. While our political beliefs may not always be epistemically justified, they can still serve important social, psychological, and practical functions. — [view in context](https://hyp.is/NzcjmgM-EfCPnGs_nGTqkg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > political beliefs can function as social commitments rather than as attempts to model the world accurately — [view in context](https://hyp.is/Zv99ygM-EfC7LXNqCTPFCQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > As fellow substacker Dan Williams has argued, many poorly justified political beliefs can be highly socially adaptive. This helps explain why deeply held political beliefs often appear unresponsive to evidence: they are shaped more by identity, group loyalty, and emotion than by detached reasoning — [view in context](https://hyp.is/aZk6lAM-EfCHmysm9d8AuA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Paul Bloom makes a similar point. He argues that while having justified or knowledgeable beliefs about everyday matters—like cooking or interpersonal relationships—has clear practical consequences, unjustified political beliefs often carry little to no direct cost for the individual. If believing something false strengthens one’s social ties, reinforces group identity, or provides psychological comfort, then truth-tracking may not be the primary function of the belief in the first place. (I have argued for this view here.) — [view in context](https://hyp.is/s1dVdgM-EfCt71tbvLO8YA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > As David Papineau argues, in cases where false beliefs are socially valuable, epistemic justification may not be the relevant standard by which to assess them. If political beliefs function as social signals rather than attempts at objective truth, then it may be a mistake to judge them purely by epistemic norms of evidence and justification. — [view in context](https://hyp.is/yYX4HgNKEfCX8X-bA5DQjA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > This perspective challenges Roeber’s claim that unjustified beliefs ought to be abandoned. If political beliefs serve purposes beyond truth-seeking—such as fostering solidarity, navigating social life, or motivating political action—then even those that lack epistemic justification might still have instrumental value. — [view in context](https://hyp.is/U0zTsANLEfCs77_tX7O3vA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Instead, we might rethink what it means to hold a belief in the political domain, recognizing that their role is often more complex than simply tracking the truth. — [view in context](https://hyp.is/WcaOtgNLEfCtd5tM01aa7A/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Moreover, political experts themselves are not immune to cognitive biases, ideological influences, and professional incentives that shape their interpretations. Because of these epistemic limitations, Roeber contends that expertise in politics does not grant individuals reliable, overarching knowledge of contested political questions. — [view in context](https://hyp.is/d45uoANLEfCtgOtJ73jwgQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Even political scientists, economists, and policy analysts must rely on contested data, partisan sources, and competing methodologies that prevent clear answers. Political issues rarely have straightforward empirical solutions. Instead, they are complex webs of normative debates, conflicting models, and unpredictable social dynamics. — [view in context](https://hyp.is/d530kgNLEfCNItdewWTjuw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > In summary, Roeber’s argument is radical in at least three ways. First, he does not just argue that ordinary people lack political knowledge but that no one possesses it, including political experts. Second, his skepticism goes beyond knowledge to justification. It’s not just that we fail to know political truths; we also lack justified beliefs about them. Third, Roeber argues that such ignorance should lead us to abandon most of our political beliefs. Recall that he says, “We should hold very few political beliefs.” Unlike more moderate forms of skepticism that call for a reduction in confidence, his position suggests that we are not even entitled to hold most political beliefs. — [view in context](https://hyp.is/RTHvOgNMEfCuIKNIc7t_Nw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Roeber rejects political agnosticism. Even though most political beliefs fall short of knowledge and even justification, he does not think we should abandon political reasoning entirely. Instead, says we can make “educated guesses.” An educated guess, as Roeber understands it, is a claim in which you’re rationally entitled to have some confidence—i.e., better than chance but well short of certainty. (To me, this sounds like a justified belief, but let’s set that aside.) You can make an educated guess only when “you’ve really done your homework.” — [view in context](https://hyp.is/aGxaMANMEfCJ5UMmaZQH4g/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > For Roeber, the problem is that most people do not make genuinely educated guesses. Instead, they form political beliefs casually and irresponsibly, through partisan media, social circles, and uncritical reliance on testimony. But for those willing to put in the intellectual work, educated guesses provide a middle ground between reckless confidence and total agnosticism. Rather than abandoning politics, Roeber’s skepticism calls for a more cautious, self-aware, and intellectually humble approach to political belief and action. Nevertheless, we should always keep in mind that “we really are just guessing.” — [view in context](https://hyp.is/qYA3YgNMEfCgXw_-x1jhaQ/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Roeber himself acknowledges that individuals have very little influence in politics and that an individual vote often makes “literally no difference.” If that’s true, then why not say that the stakes for any individual’s political belief are also very low? — [view in context](https://hyp.is/1w12NgNMEfCfQe8T96ve1w/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > In high-stakes situations—where being wrong carries serious consequences—more evidence is required for a belief to be justified. But in low-stakes situations, a person can justifiably hold a belief with relatively little evidence. (I mentioned this idea in my last post on political knowledge.) — [view in context](https://hyp.is/1x4LfANMEfCtUnsmCTrdZg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Roeber overlooks an important insight from epistemology: the idea of pragmatic encroachment. (Ironically, this is a view that Roeber himself has defended in previous work.) According to this view, whether a person is justified in believing something can depend not just on the quality of their evidence but also on the practical stakes involved — [view in context](https://hyp.is/1ytLFgNMEfC3DN8BuUP55Q/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Consider how people form opinions about the best movie actor of all time. Few people have rigorously analyzed every performance in film history, yet holding a confident belief on this subject is perfectly reasonable, given the minimal consequences of being wrong. The same logic might apply to politics. If an individual’s political belief or vote is unlikely to meaningfully impact policy, then the cost of error is low, making it rationally permissible to hold and act on beliefs formed with limited evidence. — [view in context](https://hyp.is/5yAAKgNMEfCqGOM7s7wABw/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Let’s now turn to the second challenge for Roeber’s view, which concerns the broader collective effects of political humility. Roeber argues that the best response to widespread political ignorance is a kind of humble but real engagement, one where individuals recognize their epistemic limitations and adopt a cautious, less confident approach to political belief. But even if this is epistemically reasonable at the individual level, Roeber does not fully consider the wider social context. Dogmatism, bias, and adversarial political engagement may actually produce epistemic benefits at the collective level. — [view in context](https://hyp.is/IuRQUgNNEfCP2HuZ97MF3Q/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > In The Enigma of Reason, Dan Sperber and Hugo Mercier argue that many of the cognitive biases that distort individual reasoning—confirmation bias, motivated reasoning, and overconfidence—are not merely epistemic flaws but adaptive features of human cognition. While these biases lead individuals to defend their beliefs dogmatically and resist counterevidence, they also fuel a broader process of adversarial deliberation. When opposing sides engage in debate, biased reasoning on both sides can create a productive tension, forcing each side to refine their arguments, expose weak reasoning, and respond to challenges they might otherwise ignore. — [view in context](https://hyp.is/K0VpFgNNEfCueddz_tJOAA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > In other words, what looks like poor reasoning at the individual level can, in the right social conditions, enhance the quality of collective reasoning. — [view in context](https://hyp.is/L_IzfANNEfC7V1Mxe57Xew/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > This raises a potential conflict between individual rationality and collective epistemic benefits. At the individual level, Roeber is right that people have good epistemic reasons to be more humble—they often lack justification for their political beliefs, and overconfidence leads to epistemic mistakes. But at the collective level, widespread political humility may actually undermine democratic discourse by reducing the level of engagement, weakening adversarial debate, and diminishing the competitive process through which better arguments and policies emerge. — [view in context](https://hyp.is/VMXgBANNEfCKdLfCP9AzAA/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) > Thus, Roeber’s argument for political humility may neglect an important tradeoff: while excessive confidence in one’s own views can be epistemically risky, a society in which everyone is humble may lose out on the epistemic benefits of robust political contestation. If politics is, in part, a battleground of ideas, then some degree of conviction—even overconfidence—may be necessary for the system as a whole to function effectively. — [view in context](https://hyp.is/dRd4wgNNEfCUsz9fRs72mg/michaeljhannon.substack.com/p/do-we-know-anything-about-politics) ⬆️ I am tempted to agree with this, but it does not, for now, seem to correlate with what we're seeing around us.